Shadows and photon spheres with spherical accretions in the four-dimensional Gauss-Bonnet black hole

We investigate the shadows and photon spheres of the four-dimensional Gauss-Bonnet black hole with the static and infalling spherical accretions. We show that for both cases, the shadow and photon sphere are always present. The radii of the shadow and photon sphere are independent of the profiles of accretion for a fixed Gauss-Bonnet constant, implying that the shadow is a signature of the spacetime geometry and it is hardly influenced by accretion in this case. Because of the Doppler effect, the shadow of the infalling accretion is found to be darker than that of the static one. We also investigate the effect of the Gauss-Bonnet constant on the shadow and photon sphere, and find that the larger the Gauss-Bonnet constant is, the smaller the radii of the shadow and photon sphere will be. In particular, the observed specific intensity increases with the increasing of the Gauss-Bonnet constant.


Introduction
The Event Horizon Telescope (EHT) collaboration has recently obtained an ultra high angular resolution image of the accretion flow around the supermassive black hole in M87* [1,2,3,4,5,6]. The image shows that there is a dark interior and a bright ring surrounding the dark interior, which are called the black hole shadow and photon ring, respectively. The shadow of a black hole is caused by gravitational light deflection [7,8,9,10,11]. More specifically, when light emitting from the accretion pass through the vicinity of the black hole toward the observer, its trajectory will be deflected. The intensity of the light observed by the distant observer differs accordingly, leading to a dark interior and bright ring. So far the shadows of various black holes have been investigated, and it is generally known that the shadows of spherically symmetric black holes are round and that of rotating black holes is not precisely round but deformed.
Since the release of the image and data by EHT, its various implications have been explored. For instance, the extra dimensions could be determined from the shadow of M87* [12,13], where a rotating braneworld black hole was considered. The shadows of high-redshift supermassive black holes may serve as the standard rulers [14], whereby the cosmological parameters can be constrained. The black hole companion for M87* can also be constrained through the imaging released by EHT [15]. Moreover, the information given by EHT can be used to impose constraints on particle physics via the mechanism of superradiance [16,17]. In particular, for the vector boson, it may constrain some of the fuzzy dark matter parameter space. In addition, dense axion cloud also be induced by rapidly rotating black holes through superradiance [18].
Accretion matters are apparently important for the shadows of black holes, because many astrophysical black holes are believed to be surrounded by accretion matters. The first image of a black hole surrounded by an emitting accretion thin disk was pictured out in [19]. For a geometrically thin, optically thick accretion disk [20], it was found that the mass of the disk would affect the shadow of a black hole and as the mass grows, the shadow becomes more prolate. In particular, by reanalyzing the trajectory of a light ray, the shadow of a Schwarzschild black hole with thin and thick accretion disks have been clarified and detailed recently [21]. It is found that there exists not only the photon ring 1 but also the lensing ring, and the lensing ring makes a significant contribution to the observed flux while the photon ring makes little. In addition, the observed size of the central dark area is found to be determined not only by the gravitational redshift but also by the emission profile. When the accretion matter is spherically symmetric, there is also a shadow for a black hole [22]. The location of the shadow edge is found to be independent of the inner radius at which the accreting gas stops radiating [23]. The size of the observed shadow can serve as a signature of the spacetime geometry, because it is hardly influenced by accretion details. This result is different from the case in which the accretion is a disk [21].
In this paper, we intend to investigate the shadow of a four-dimensional Gauss-Bonnet black hole with spherical accretions [24]. The Gauss-Bonnet term in the Lagrangian is topologically invariant in four dimensional spacetime. Thus in order to consider the dynamical effect of Gauss-Bonnet gravity, one is generically required to work in higher dimensions [25,26]. But very recently, Glavan and Lin has proposed a general covariant Gauss-Bonnet modified gravity in four dimension by rescaling the Gauss-Bonnet coupling constant α → α/(D − 4) and taking the limit D → 4 [24]. Such a theory can bypass Lovelock's theorem and avoid Ostrogradsky instability 2 . Many characteristics of the fourdimensional Gauss-Bonnet black hole have been investigated [27,28,29,30,31,32,33,34,35,36,37,38,39,40,41]. In particular, gravitational lensing in ordinary medium [46] and homogeneous plasma [47] by black holes in four-dimensional Gauss-Bonnet gravity have been explored. The shadows cast by the spherically symmetric [48] and rotating [49] four-dimensional Gauss-Bonnet black hole have also been studied, where nevertheless the details the shadows are not figured out except the radius. It will be more interesting to investigate the corresponding light intensity of the shadow, which comprises the main issue of this paper. To be more precise, in this paper, we are interested in the spherical accretions, which can be classified into the static and infalling one. On the one hand, we want to explore how the Gauss-Bonnet constant affects the radii of the shadow and photon sphere as well as the light intensity observed by a distant observer. On the other hand, we want to explore how the dynamics of the accretion affects the shadow of the black hole. As a result, we find that the larger the Gauss-Bonnet constant is, the smaller the radii of the shadow and photon sphere will be, and the larger the intensity will be. In addition, the shadow of the infalling accretion are found to be darker than that of the static one because of the Doppler effect.
The remainder of this paper is organized as follows. In Section 2, we investigate the motion of the light ray near the four-dimensional Gauss-Bonnet black hole and figure out how it is deflected. In Section 3, we investigate the shadows and photon spheres with the static spherical accretion. To explore whether the dynamics of the accretion affect the shadow and photon sphere, the accretion is supposed to be infalling in Section 4. Section 5 is devoted to the conclusions and discussions. Throughout this paper, we will set G = = c = k B = 1.

Light deflection in the four-dimensional Gauss-Bonnet black hole
Starting from the following Einstein-Hilbert action with an additional Gauss-Bonnet term rescaling the Gauss-Bonne coupling constant α → α/(D − 4) and taking the limit D → 4, one can obtain the four-dimensional spherically symmetric Gauss-Bonnet black hole as with where M is the mass of the black hole. Note that the same solution was already found previously in [50] by considering the Einstein gravity with Weyl anomaly. Solving the equation F (r) = 0, one can obtain two solutions, in which r + and r − correspond to the outer horizon (event horizon) and inner horizon, respectively. To assure the existence of a horizon, the Gauss-Bonnet coupling constant should be restricted in the range −8 ≤ α/M 2 ≤ 1. For the case α > 0, there are two horizons, and for the case α < 0 there is only a single horizon. In order to investigate the light deflection caused by the four-dimensional Gauss-Bonnet black hole, we should find how the light ray moves around the black hole. As we know, the light ray satisfies the geodesic equation, which can be encapsulated in the following Euler-Lagrange equation d dλ with λ the affine parameter,ẋ µ the four-velocity of the light ray and L the Lagrangian, taking the form as As in [7,8,9], we focus our attention on the light ray that moves on the equatorial plane, i.e., θ = π 2 andθ = 0. In addition, because none of the metric coefficients depends explicitly on time t and azimuthal angle φ, there are two corresponding conserved quantities, E, L. Combining Eqs.(3), (5) and (6), the time, azimuthal and radial component of the fourvelocity can be expressed aṡ where we have redefined the affine parameter λ → λ/|L|, and b = |L| E , which will be called the impact parameter thereafter. The + and − in Eq. (8) correspond to the light ray traveling in counterclockwise and clockwise respectively along azimuthal direction. Eq. (9) can also be rewritten asṙ where is an effective potential. The conditions for the photon sphere orbit areṙ = 0 andr = 0, which can be translated to where the prime denotes the first derivative with respect to the radial coordinate r. Based on this equation, we can obtain the radius and impact parameter for the photon sphere, which are shown together with the size of the event horizon in Table 1 for different α. From this table, we know that not only the radius of the event horizon, but also the radius and impact parameter for the photon sphere decreases as one increases α. Table 1. The radius r ph , impact parameter b ph of the photon sphere and the event horizon r + for different α with M = 1. Here we would like to take α = −5.5, 0.555 as two examples with the corresponding effective potential depicted in Figure 1. We can see that at the event horizon, the effective potential vanishes. It increases and reaches a maximum at the photon sphere, and then decreases as the light ray moves outwards. As a light ray moves in the radially inward direction, the effective potential will affect its trajectory. In Region 1, the light will encounter the potential barrier and be reflected back in the outward direction. In Region 2, namely b = b ph , the light will asymptotically approach the photon sphere. Since the angular velocity is non-zero, it will revolve around the black hole infinitely many times. In Region 3, the light will continue moving in the inward direction since it does not encounter the potential barrier. Eventually it will enter the inside of the black hole through the event horizon.
Furthermore, the trajectory of the light ray can be depicted according to the equation of motion. Combining Eqs. (8) and (9), we have   In order to integrate it out conveniently, we set u = 1/r. Accordingly Eq.(13) becomes With Eq. (14), we can obtain φ with respect to u. Employing the ParametricPlot 3 , we can plot the trajectory of the light ray, which is shown in Figure 2. The black, red and green line correspond individually to b < b ph , b = b ph and b > b ph . As one can obviously see, For the case b < b ph , the light ray drops all the way into the black hole, which corresponds to Region 3 in Figure 1. For the case b > b ph , the light ray near the black hole is reflected back, which corresponds to Region 1 in Figure 1. And for the case b < b ph , the light ray revolves around the black hole, which corresponds to Region 2 in Figure 1. Note that for b > b ph , in order to plot the geodesic, we should find a turning point, where the light ray changes its radial direction. The turning point is determined by the equation G(u) = 0, where G(u) has been defined in Eq. (14).

Shadows and photon spheres with rest spherical accretion
In this section, we intend to investigate the shadow and photon sphere of the fourdimensional Gauss-Bonnet black hole with static spherical accretion, which is assumed to be optically thin. To this end, we should find the specific intensity observed by the observer (ergs −1 cm −2 str −1 Hz −1 ). The observed specific intensity at the observed photon frequency ν o can be found integrating the specific emissivity along the photon path [51,52] where g = ν o /ν e is the redshift factor, ν e is the photon frequency, dl prop is the infinitesimal proper length, and j(ν e ) is the emissivity per unit volume measured in the rest frame of the emitter.
In the four-dimensional Gauss-Bonnet black hole, g = F (r) 1/2 . Concerning the specific emissivity, we also assume that it is monochromatic with rest-frame frequency ν r , that is According to Eq.(2), the proper length measured in the rest frame of the emitter is  in which dφ/dr is given by Eq. (13). In this case, the specific intensity observed by the infinite observer is The intensity is circularly symmetric, with the impact parameter b of the radius, which satisfies b 2 = x 2 + y 2 .
Next we will employ Eq. (18) to investigate the shadow of the four-dimensional Gauss-Bonnet black hole with the static spherical accretion. Note that the intensity depends on the trajectory of the light ray, which is determined by the impact parameter b. So we will investigate how the intensity varies with respect to the impact parameter. For different α, the numeric result is shown in Figure 3. From this figure, we can see that the intensity increases rapidly and reaches a peak at b ph , and then drops to lower values with increasing b. This result is consistent with Figure 1 and Figure 2. Since for b < b ph , the intensity originating from the accretion is absorbed mostly by the black hole. And for b = b ph , the light ray revolves around the black hole many times, so that the observed intensity is maximal. While for b > b ph , only the refracted light contributes to the intensity of the observer. As b becomes larger, the refracted light becomes less. The observed intensity thus vanishes for large enough b. In principle, the peak intensity at b = b ph should be infinite because the light ray revolves around the black hole infinite times and collect an arbitrarily large intensity. Because of numerical limitations and the logarithmic form of the intensity, the actual computed intensity never goes to infinity, which has also been well addressed in [21,23].
From Figure 3, we also can observe how the Gauss-Bonnet coupling constant affects the observed intensity. For all the b, the larger the coupling constant is, the stronger the intensity will be.
The shadow cast by the four-dimensional Gauss-Bonnet black hole in the x, y plane is shown in Figure 4. We can see that outside the black hole, there is a bright ring, which is the photon sphere. The radius of the photon sphere for different α have been listed in Table 1. Obviously, the result in Figure 4 is consistent with that in Table 1. That is, the larger the Gauss-Bonnet constant is, the smaller the radius of the photon sphere is. Moreover, we can see that inside the shadow, the intensity does not go to zero but has a small finite value. The reason is that part of the radiation has escaped to infinity. For r > r ph , the solid angle of the escaping rays is 2π(1 + cos θ), and for r < r ph , the solid angle of the escaping rays is 2π(1 − cos θ), where θ is given by by only counting the escaping light rays, we have the net luminosity observed at infinity as  For different α, the numeric result is listed in Table 2. We can see that the net luminosity increases with increasing α. For the Schwarzschild black hole, the net luminosity is found to be L ∞ = 0.32 [23]. Obviously, for the positive α, the net luminosity in the four-dimensional  Gauss-Bonnet black hole is larger than the Schwarzschild black hole, and for the negative α, the net luminosity in this spacetime is smaller than that of Schwarzschild black hole.

Shadows and photon spheres with infalling spherical accretion
In this section, we allow the optically thin accretion to move towards the black hole. This model is thought to be more realistic than the static accretion model since most of the accretions are mobile in the universe. For simplicity, we assume that the accretion free falls on to the black hole from infinity. We still employ Eq. (18) to investigate the shadow of the four-dimensional Gauss-Bonnet black hole. Different from the static accretion, the redshift factor for the infalling accretion should be evaluated from in which k µ =ẋ µ is the four-velocity of the photon, u µ o = (1, 0, 0, 0) is the 4-velocity of the distant observer, while u µ e is the 4-velocity of the accretion under consideration, given by The four-velocity of the photon has been obtained previously in Eq.(7)-Eq.(9). We know that k t = 1/b is a constant, and k r can be inferred from k γ k γ = 0, that is where the sign +(−) correspond to the case that the photon gets close to (gets away from) In addition, the proper distance can be defined by where λ is the affine parameter along the photon path γ. Concerning the specific emissivity, we also assume that it is monochromatic, so that Eq.(16) can be used. The intensity in Eq.(15) thus can be expressed as Now we will use Eq. (26) to investigate the shadow of the black hole numerically. For different α, the intensity with respect to b observed by the distant observer are shown in Figure 5. Similar to the static accretion, we find that the intensity increases towards larger b, reaches a maximum intensity at b = b ph , and then drops away. We can also observe the effect of α on the intensity. The larger the value of b is, the larger the observed intensity will be.
The 2-dimensional image seen by a distant observer is shown in Figure 6. We can see that the radius of the shadow and the location of the photon sphere are the same as the static accretion. A major new feature is that in the central region, the shadow with infalling accretion is darker than that with the static accretion, which is well accounted for by the Doppler effect. Nearer the event horizon of the black hole, this effect is more obvious.
It has been argued that in the universe, the accretion flows do have inward radial velocity, and the velocity tends to be large precisely at the radii of interest for the shadow formation. Therefore the model with radially infalling gas is most appropriate for comparison with the image of M87*.
To explore how the profile of the specific emissivity affects the shadow of the black hole, we here choose different profiles of j(ν e ), and the corresponding intensities are shown in Figure 7. From this figure, we can see obviously that the intensity in these cases has the behavior similar to the case j(ν e ) = 1/r 2 . That is, the peak is always located at b = b ph . The difference is that the intensity decays faster for the higher power of 1/r, which makes the peak more obvious. This phenomenon is more prominent in the 2-dimensional image shown in Figure 8.
Our result in Figure 7 and Figure 8 show that though the profile of the spherical accretion affects the intensity of the shadow, it does not affect the characteristic geometry such as the radius of the shadow, which is determined only by the geometry of the spacetime.

Discussions and conclusions
In this paper, we have investigated the shadows and photon spheres cast by the fourdimensional Gauss-Bonnet black hole. We first obtain the radius of the photon sphere and critical impact parameter for different Gauss-Bonnet constants, and find that the larger the Gauss-Bonnet constant is, the smaller the radius of the photon sphere and critical impact parameter will be, which is consistent with the previous results [48,49].
More importantly, we obtain the specific intensity I(ν o ) observed by a distant observer, in which the accretion was supposed to be either static or infalling. For both cases, we find that the specific intensity increases with the increase Gauss-Bonnet constant. We plot  the shadow in the x − y plane, and find that there is a bright sphere ring outside the dark region. The center region of the shadow with the infalling accretion turns out to be darker than that with the static accretion due to the Doppler effect. We also investigate the effect of the profile of the accretion on the shadow. As a result, it is found that though the profile affect the intensity of shadow, it does not affect the characteristic geometry such as the radius of the photon sphere. In Ref. [23], the emission originating from the accretion was cut-off at different locations, the size of the shadow was found to be independent of the locations. Obviously, our result is consistent with this observation in Ref. [23].
The EHT Collaboration has molded M87* with the Kerr black hole, and claimed that the observation supports the General Relativity. In this paper, we did not consider the Kerr-like black hole in the four-dimensional Gauss-Bonnet gravity. But as stressed in Ref. [51], the spherically symmetric black hole can produce qualitatively similar results. More broadly, the simplified spherical model captures key features that also appear in state of the art general-relativistic magnetohydrodynamics models [5], whether they are spinning or not.
In addition, the real accretion flows are generically not spherically symmetric The hot accretion flow in M87* and most other galactic nuclei consists of a geometrically thick, quasi-spherical disk. It will be more interesting to investigate the shadow with a thick disk accretion. Recently, Ref. [21] has investigated the shadow with a thin and thick accretion. They reanalyze the orbit of photon and redefine the photon ring and lensing ring, in which the lensing ring is the light ray that intersects the plane of the disk twice and the photon ring is that intersects the plane three or more times. They define a total number of orbits as n ≡ φ/2π. In this case, n > 3/4 corresponds to the light ray cross the equatorial plane at least twice, n > 5/4 corresponds to the light ray cross the equatorial plane at least three times, and n < 3/4 corresponds to the light ray cross the equatorial plane once. For the case α = 0.111, the trajectory of the light ray is shown in Figure 9. Compared it with Figure 2, we can see that the photon ring is around the photon sphere, and the lensing ring is around the photon ring. It will be interesting to investigate the shadow, photon ring, and lensing ring with a thin or thick disk in the four-dimensional Gauss-Bonnet black hole.